An ABC-Optimized Multimodal Intelligent Adaptive Control Strategy for Six DOF Wave Compensation

Ship-borne Stewart platforms ensure safe marine transportation by compensating for wave-induced 6-DOF motions, but face challenges: 1) strong nonlinear wave interference introduces significant time-varying uncertainty; 2) abrupt load perturbations cause parameter fluctuations, degrading tracking precision; 3) complex interference environments reduce control robustness. To address these, an artificial bee colony (ABC) algorithm-optimized fuzzy radial basis function neural network PI control strategy (ABC-FNN-PI) is proposed. This approach integrates ABC optimization, fuzzy logic, and RBF neural networks, overcoming empirical parameter-tuning limitations and single-controller flaws. The controller exploits fuzzy inference to provide fast response and robust adaptation without relying on an accurate model; to overcome the inherent limitations of rule-based regulation and further improve compensation accuracy under complex nonlinear, an RBF network is integrated to perform data-driven nonlinear refinement, enabling multimodal collaboration and intelligent adaptability. The results under Sea States 3–5 with superimposed random measurement noise and abrupt load perturbations indicate that, compared with conventional PI, Fuzzy PI, and active disturbance rejection control (ADRC), the proposed fuzzy-inference-driven and neural-refined multimodal controller achieves consistently higher compensation accuracy and stability, and exhibits strong robustness against wave disturbances, load variations, and measurement noise, offering a reliable solution for high-precision motion control in harsh marine environments.